Predicting New Thermoelectric Materials Through Automated Computer Process: New Study Reveals
In an exciting breakthrough, an international research team led by Andrej Pustogow from TU Wien has predicted new materials for generating "green energy" using an automated computer protocol. Their findings, published in the journal Science Advances in 2025, have identified nickel-germanium (NiGe) as a highly promising thermoelectric material.
The knowledge taught in schools and universities is limited, and textbooks only reflect the state of knowledge at the time of publication. This can lead to a long search for materials with specific properties in the wrong place. Recognising this, Pustogow stated that a new idea was needed that goes beyond textbook knowledge in the field of thermoelectricity.
The researchers selected transition metals iron, cobalt, and nickel as their starting point. Their approach was to determine the relevant material properties using automated calculations in a well-defined part of the periodic table. The more than 100 elements in the periodic table are too many for a blind search due to the excessive computing time for combinations of just a few dozen types of atoms.
The team's approach allowed them to achieve the same result with minimal material and time expenditure compared to traditional materials science methods. Ab initio simulations and computational screening enabled the prediction of promising candidate materials by analysing electronic structures and transport properties before synthesis.
Simulations revealed how energy filtering can enhance the thermoelectric power factor in NiGe by selectively scattering low-energy charge carriers while allowing high-energy carriers to contribute, improving efficiency. The material with the best predicted properties was the compound of nickel and tin, but it is difficult to produce under normal conditions in the laboratory, so the focus was on NiGe.
Samples of nickel and germanium ultimately demonstrated excellent thermoelectric properties in the laboratory, as predicted by the computer protocol. The material with the best performance was Ni₃Ge, which was subsequently synthesized and experimentally confirmed to have outstanding thermoelectric performance.
Tech companies such as Google and Microsoft are also pursuing this computer-aided design approach using artificial intelligence. The combined computational-experimental strategy significantly accelerates discovering new thermoelectric materials by focusing on promising compounds predicted with high confidence rather than relying on trial-and-error laboratory experimentation alone.
The study's DOI is 10.1126/sciadv.adv7113. Fabian Garmroudi, first author of the study, mentioned that Abram Ioffe suggested semiconducting materials for thermoelectric applications in the 1930s. Despite intensive research on semiconductors for nearly a century, widespread applications have not been achieved. However, with the advancements in computer simulations and targeted experimental approaches, the future of thermoelectric materials looks promising.
References: 1. [Article Title], [Journal Name], [Year] 2. [Article Title], [Journal Name], [Year] 3. [Article Title], [Journal Name], [Year] 4. [Article Title], [Journal Name], [Year]
Science and technology have played significant roles in the recent breakthroughs in generating 'green energy' through thermoelectric materials. While traditional materials science methods can be time-consuming and costly, the research team led by Andrej Pustogow, using an automated computer protocol, successfully predicted promising candidate materials like nickel-germanium (NiGe) with minimal effort. Furthermore, tech companies like Google and Microsoft are adopting this computer-aided design approach using artificial intelligence to accelerate the discovery of new thermoelectric materials, focusing on those predicted with high confidence rather than relying solely on trial-and-error experimentation.
